# QuoTA: Query-oriented Token Assignment via CoT Query Decouple for Long Video Comprehension

arXiv: https://arxiv.org/abs/2503.08689 · HTML: https://mlsanity.com/paper/2503.08689/

## Key facts

- Cited by 21 papers in the map (data through 14 Oct 2026).
- Placed in Visual token pruning in multimodal models, one of 12 directions in Efficient LLM inference and serving.
- First posted on arXiv 11 Mar 2025.

## Placed in

- Efficient LLM inference and serving / [Visual token pruning in multimodal models](https://mlsanity.com/inference/visual-token-pruning-in-multimodal-models/)

## Built on by (most cited first, 21 of 21)

- [HoliTom: Holistic Token Merging for Fast Video Large Language Models](https://mlsanity.com/paper/2505.21334/) (2025)
- [Video-MMLU: A Massive Multi-Discipline Lecture Understanding Benchmark](https://mlsanity.com/paper/2504.14693/) (2025)
- [VideoITG: Multimodal Video Understanding with Instructed Temporal Grounding](https://mlsanity.com/paper/2507.13353/) (2025)
- [A Survey of Token Compression for Efficient Multimodal Large Language Models](https://mlsanity.com/paper/2507.20198/) (2025)
- [Flow caching for autoregressive video generation](https://mlsanity.com/paper/2602.10825/) (2026)
- [FlexSelect: Flexible Token Selection for Efficient Long Video Understanding](https://mlsanity.com/paper/2506.00993/) (2025)
- [See&Trek: Training-Free Spatial Prompting for Multimodal Large Language Model](https://mlsanity.com/paper/2509.16087/) (2025)
- [Less Is More, but Where? Dynamic Token Compression via LLM-Guided Keyframe Prior](https://mlsanity.com/paper/2512.06866/) (2025)
- [SocialOmni: Benchmarking Audio-Visual Social Interactivity in Omni Models](https://mlsanity.com/paper/2603.16859/) (2026)
- [APVR: Hour-Level Long Video Understanding with Adaptive Pivot Visual Information Retrieval](https://mlsanity.com/paper/2506.04953/) (2025)
- [Event-Anchored Frame Selection for Effective Long-Video Understanding](https://mlsanity.com/paper/2603.00983/) (2026)
- [Enhancing Long Video Question Answering with Scene-Localized Frame Grouping](https://mlsanity.com/paper/2508.03009/) (2025)
- [ViKey: Enhancing Temporal Understanding in Videos via Visual Prompting](https://mlsanity.com/paper/2603.23186/) (2026)
- [WaveZip: Wavelet-Driven Space-Time Decoupling for Video Token Condensation](https://mlsanity.com/paper/2607.23265/) (2026)
- [ChronoForge-RL: Chronological Forging through Reinforcement Learning for Enhanced Video Understanding](https://mlsanity.com/paper/2509.15800/) (2025)
- [SpecEyes: Accelerating Agentic Multimodal LLMs via Speculative Perception and Planning](https://mlsanity.com/paper/2603.23483/) (2026)
- [OmniScope: Modality-Decoupled Token Compression for Omnimodal Large Language Models](https://mlsanity.com/paper/2607.23193/) (2026)
- [One Ranking, Any Budget: Matryoshka Evidence-to-Context Frame Selection for Long-Video Understanding](https://mlsanity.com/paper/2608.05707/) (2026)
- [Query-Aware Token Budgeting for Efficient Late-Interaction Visual Document Retrieval](https://mlsanity.com/paper/2609.07262/) (2026)
- [KeyRec: Bounded Visual Memory for Streaming and Long-Video Understanding](https://mlsanity.com/paper/2609.32182/) (2026)
- [VideoLoop: Looped Working Memory Against Semantic Thrashing in Long-Form Video Agents](https://mlsanity.com/paper/2609.38119/) (2026)

## Abstract (arXiv)

Recent advances in long video understanding typically mitigate visual redundancy through visual token pruning based on attention distribution. However, while existing methods employ post-hoc low-response token pruning in decoder layers, they overlook the input-level semantic correlation between visual tokens and instructions (query). In this paper, we propose QuoTA, an ante-hoc training-free modular that extends existing large video-language models (LVLMs) for visual token assignment based on query-oriented frame-level importance assessment. The query-oriented token selection is crucial as it aligns visual processing with task-specific requirements, optimizing token budget utilization while preserving semantically relevant content. Specifically, (i) QuoTA strategically allocates frame-level importance scores based on query relevance, enabling one-time visual token assignment before cross-modal interactions in decoder layers, (ii) we decouple the query through Chain-of-Thoughts reasoning to facilitate more precise LVLM-based frame importance scoring, and (iii) QuoTA offers a plug-and-play functionality that extends to existing LVLMs. Extensive experimental results demonstrate that implementing QuoTA with LLaVA-Video-7B yields an average performance improvement of 3.2% across six benchmarks (including Video-MME and MLVU) while operating within an identical visual token budget as the baseline. Codes are open-sourced at https://github.com/MAC-AutoML/QuoTA.